Columbia/Kobal Collection

Jimmy Stewart in the 1939 film Mr. Smith Goes to Washington

The United States Senate, wrote chief New York Times congressional correspondent William S. White in 1957, is “a body that never wholly changes and never quite dies,” a place “where the national past and the national future meet and soundlessly merge.” White wrote those words when the Senate was on the brink of a revolution: the next three elections would bring roughly fifteen outright liberals to the chamber, adding to the handful already there. This irrevocably shifted the balance of power away from the Southern conservatives, creating the conditions that made possible the passage of the Great Society legislation. Fifty-three years later, though, White’s description is again becoming all too true. The Senate, except for a few brief moments such as the burst of activity in the 1960s and 1970s, has been where progressive legislation goes to die—not always soundlessly, but almost always.

The passage of the health care reform bill was an anomaly, a product of the fact that for the brief period between Al Franken’s certification as the junior senator from Minnesota (he was sworn in July 7, 2009) and Scott Brown’s as the new senator from Massachusetts almost exactly seven months later, the Democrats had the sixty votes needed to invoke “cloture,” i.e., cut off debate and bring any measure to the floor. Today, as midterm elections loom in which Republicans anticipate substantial gains, we can see clearly that there will not be sixty votes for a controversial piece of legislation for some time.
The Senate’s version of the financial reform bill squeaked by in May with the backing of four Republicans from blue states (Scott Brown, Chuck Grassley of Iowa, and Olympia Snowe and Susan Collins of Maine). Though it contained some strong provisions, the Senate bill was not nearly as forceful as liberals had hoped on matters like requiring banks to spin off their proprietary trading operations. In late July, Democrats acknowledged officially what had been clear for several months—that no major energy bill would make it through the Senate this year, leaving the House of Representatives’ bill, passed the previous June, to expire, with no prospect that the matter will be taken up in the near future.

And when Congress returns to work next January, the Senate will likely have five or six or even eight fewer Democrats, with an outside possibility that they will lose control of the body altogether. Major progressive legislation will become impossible. Even if President Obama wins a second term and the Democrats retain control in the House of Representatives and maintain their hold on fifty-four or fifty-five seats in the Senate—a substantial majority, in other words—they will be able to pass only contingent and watered-down versions of their programs. In politics, we normally consider 55 percent a commanding majority, even a landslide; but in the Senate, having 55 percent of the votes usually gets you nothing.

Many will say at this point that it was always this way. No, it was not. While it’s true that the Senate has been more conservative and even reactionary than not throughout its history, notably with regard to the way it blocked civil rights legislation for many years, the fact is that obstructionism is empirically worse today than ever, or at least since 1917, when the current “cloture” system was first adopted. The idea that the Senate is supposed to behave as it is behaving today is one of several myths (or outright lies) surrounding the “world’s greatest deliberative body.”

The truth is that no institution of American government is more responsible for our inability to address pressing national problems than the Senate, and no institution is in greater need of reform. Another truth, alas: probably no institution is more resistant to reform. Nevertheless, Senate Majority Leader Harry Reid is trying, having placed filibuster reform on the Senate’s agenda, with the possibility that votes to change the procedure may be taken next January. “The filibuster has been abused,” Reid said in March. “Next Congress, we’re going to take a look at it.”1 This process has begun, with three hearings on possible reforms.

While all this is of concern to liberals at the moment, the problem of the Senate should trouble all citizens. Any party with a president and fifty-nine senators (counting the two independents who caucus with the party), not to mention a seventy-eight-seat House majority, ought to have a fair chance to enact its programs. It would need minority input, to be sure, but the specter of a minority veto, which some founders warned against, would not loom over all deliberations, as is the case today, thanks largely to the cloture rule. In fairness to both parties, when they have a decent-sized majority, they should not have to muster a supermajority. Chances of passing legislation without one, slim as they are, hinge on dispensing with current myths about the Senate. There are four chief myths.


The first myth is that the founders wanted the upper house to follow the principles of supermajorities, such as the sixty votes now needed to end debate on a measure. This idea is often supported by the famous statement of George Washington, who explained to Thomas Jefferson that the Senate would “cool” the inflamed passions of any given moment as a saucer cools hot tea. It’s true that this was the intention, which is why senators were elected for six years rather than two and were elected initially by state legislatures rather than directly.

But aside from limited instances, such as the expulsion of a member, it does not follow that the founders wanted supermajorities. As Sarah A. Binder and Steven S. Smith show in their book Politics or Principle? some founders—George Mason of Virginia among them—backed supermajority requirements, but many were suspicious of them. The Continental Congress, under the Articles of Confederation, had been run on the supermajority principle—most legislation needed the support of two thirds of the states, or nine out of thirteen, to pass—and the results were unsatisfying. James Madison acknowledged that “more than a majority” might be justifiable in limited instances but argued that requirements for a supermajority were open to a decisive objection:

In all cases where justice or the general good might require new laws to be passed, or active measures to be pursued, the fundamental principle of free government would be reversed. It would be no longer the majority that would rule; the power would be transferred to the minority.

Alexander Hamilton echoed this view in Federalist No. 22: “To give a minority a negative upon the majority (which is always the case where more than a majority is requisite to a decision) is…to subject the sense of the greater number to that of the lesser number.” He added that such a provision would “destroy the energy of government,” handing outsized power to “an insignificant, turbulent or corrupt junto.”2

The second myth is that the founders specifically sanctioned the filibuster. This view is advanced from time to time, mostly by conservatives, but it is simply not true. Of congressional rules, the Constitution says only this, in Article I, Section 5: “Each House may determine the Rules of its Proceedings.” As Gregory Koger points out in his impressively researched Filibustering:

The “right” to filibuster in the Senate is based on tenuous precedents and informal practices. At no point did senators consciously choose to remake their chamber or transform American politics. It just happened, and it happened so quietly we barely noticed.

Delaying tactics such as refusing to show up for a quorum began in the early 1800s, and the first filibusters, in the Mr. Smith Goes to Washington sense of holding the floor for hours on end, date to the 1830s. Filibustering was actually more common in the House of Representatives (under old rules) than in the Senate, until the Senate caught up in the 1890s with a spate of filibustering on issues such as returning to the gold standard. Binder and Smith, amusingly but probably accurately, attribute such filibusters to the fact that the renovation of the Senate chamber at the time resulted in better air circulation, permitting lengthier orations. In any case, the Constitution had nothing to do with it.

The third myth is that the Senate has consistently opposed any and all attempts to cut off debate and reform its procedures; that senators want things just the way they are. Many have and do, but there have been and are important exceptions. In the years leading up to the Civil War, Henry Clay of Kentucky and Stephen Douglas of Illinois, who represented opposing political parties—Clay was a Whig, Douglas a Democrat—concurred that debate in the Senate had gotten out of hand. Clay wanted to pass a motion to enable the Senate “to call the previous question”—that is, to permit a simple majority to force a vote on the matter at hand; which had been permissible until 1806, when the Senate deemed the rule unnecessary and let it drop. “Far from the rule being condemned,” Clay said in 1841, “it would be generally approved.”

Generally—but not wholly. In a maddening catch-22 that remains with us today, establishing the principle of simple-majority rule requires the backing of a supermajority of two thirds of the senators to make the change. A minority of senators, but enough of a minority, opposed Clay’s efforts, and the measure died. There have been numerous other instances when senators did try to move in the direction of fairness and limit the dilatory maneuverings that characterize the body. They did so by introducing in 1917 the cloture rule by which a vote of 60 percent can now close off debate; and they also did so by, for example, removing budgetary measures from the requirement for a supermajority to achieve cloture (the Byrd Rule of 1974). Precedents for reform, attempted and realized, are ample.


The final main myth is to be found in the so-called “little-harm” thesis. Defenders of the filibuster maintain that the supermajority requirement really hasn’t had much impact on the work of the Senate after all; it has killed legislation only rarely, and in several cases, its defenders maintain, it has improved legislation, making it more acceptable to the American public, cooling hot tea just as Washington envisioned.

This may be so in some instances. But typically, filibusters have put off for decades actions the nation should have taken years before—civil rights, notably, including anti-lynching laws. According to the Senate historian’s office, there have been 892 cloture votes since 1919, and 502 of them, or 56 percent, have failed. That is certainly not minimal impact on Senate business. As to the question of whether filibusters make legislation more broadly reflective of public opinion, this is difficult either to prove or disprove. Under a simple-majority requirement, obtaining the support of the fifty-first senator is decisive. Under the current rules, it is necessary to have the support of the sixtieth senator. There seems little basis for presuming that the sixtieth senator is more representative of the broader public than the fifty-first.

Illustration by Thomas Nast

If filibuster reform is to be serious, such myths must be shattered, which won’t be easy. What must be made clear is that the situation today is out of hand—that what happens now is not just “politics as usual” and that both sides do not abuse the threat of the filibuster equally.

As noted, the modern cloture system went into effect in 1917, with the adoption of Rule 22, which was intended to make it more difficult to delay or avoid a vote by stretching out debate. It was passed partially in response to public outcry against the “little band of willful men,” the senators who obstructed Woodrow Wilson’s desire to arm merchant vessels in the years before the US entered World War I. Designed to rationalize a procedure for bringing matters to the Senate floor for votes, the same process, with some refinements, is the one we follow today.

To advance to a final vote, a bill must first be placed on the Senate calendar, which mainly happens after it is voted out of a committee. Then, after it is placed on the calendar, it still must reach the Senate floor, which can happen in two ways. The first is by unanimous consent, in which senators all informally agree to bring a matter to the floor. As we saw with the dispute over extension of unemployment benefits, one senator—in that case, Kentucky Republican Jim Bunning—can stop unanimous consent entirely on his or her own. But most of the time, at least on uncontroversial matters, the Senate retains enough of its old, pre-polarization decorum to allow unanimous consent to work.

According to Rule 22, the other way for a matter to reach the floor is through the filing of a cloture motion. This is the method used on many more disputed matters. Here, a member of the majority senses or believes, by means of private communications or outright public statements, that the minority is willing to filibuster a piece of legislation. A member of the majority in such a case files a cloture motion referring to a particular bill, a petition that sixteen senators must sign. The motion is filed. It has a day to “ripen,” and the majority leader then can call the motion up on the second day. This is when the first vote takes place requiring the supermajority of sixty. If the majority fails the first time, it can try again, as we saw with financial regulatory reform, when Harry Reid didn’t get sixty votes to bring the bill to the Senate floor until the third attempt. In the first attempt, the motion to bring the reform bill to the Senate floor received only fifty-seven votes.

Matters can get much more complex than that. While all votes on final passage of a bill require just a simple majority of fifty-one, a number of procedural votes can also require sixty—for example, the naming of senators to serve on the conference committee to work out differences between its bill and the House’s. In addition, matters that have a direct impact on revenue can be passed under so-called “reconciliation” rules, which do not require a cloture vote—the 2001 Bush tax cuts were passed in that fashion and any changes to those cuts may be made by simple majority. But the process I have described of filing a cloture motion is the one that gets press attention—the one, for example, that played out in the Senate in the days before Christmas last year, when debate on the health care bill was ended (60–40) and the bill passed (60–39). Many casual observers thought this first vote was on the bill itself, not on cloture.

A look at the historical record shows that for a time, the rule worked pretty well. The period from the 1930s to the early 1970s saw a vast decline in filibustering, except, specifically, on civil rights–related matters. The querulous Huey Long, unsurprisingly, tested his fellow senators’ patience with around twenty filibusters from 1933 to 1935, on both important and trivial matters; his last filibuster put an end to a catch-all appropriation just two weeks before he was killed by an assassin. Most senators came to agree that the filibuster was heavy artillery, a long-bomb pass, to be used only in the most extreme and contentious circumstances. Thus, Strom Thurmond set a record with his twenty-four-hour speech against civil rights in August 1957 (the Senate had air-conditioning by then). In the 1970s, that consensus began to decline. Legislators in both parties were to blame, driven by the then-new forces that were changing politics: the decline of party, the defining role of money, the importance of television.

We can’t measure the incidence of filibusters in this modern era. Usually, they are merely threatened by the minority, at which point the majority begins the cloture process and tries to round up sixty votes. Actually rolling out the cots, which happens very rarely—yes, the United States Senate maintains a storeroom of cots—is considered an unacceptable delay in Senate business. But we can measure petitions for cloture, which are typically filed by majorities in response to perceived threats of filibusters. Experts agree that the number of cloture motions correlates quite closely to the number of threatened filibusters. The picture since the 1970s is an increasingly alarming one. From Rule 22’s enactment through 1970, the number of motions filed in any given Congress (a two-year period) never surpassed single digits. Then, quickly, there were twenty-three in 1971–1972, forty-four in the next Congress, and eighty in 1993–1994.

After a brief dip, we come to the present day, when the number of motions has gone through the roof, reaching 139 in the Congress of 2007–2008. So far in the current Congress, there have been only a small number of actual filibusters—for example, the attempt to delay the extension of unemployment benefits—but 117 cloture motions have been filed.3 They affect a substantial portion of the Senate’s business. The political scientist Barbara Sinclair has found that about 8 percent of major Senate bills in the 1960s were subject to filibuster politics, by which of course she means cloture motions to head off filibusters; in our decade, she puts that figure at 70 percent.4

Bearing in mind which party controlled the Senate during any given two-year period, one can glean a sense of which party threatens filibusters more often while in the minority. It is certainly true that both parties have used the threat excessively. But the numbers show that Democrats have filed more cloture petitions, suggesting that Republicans have threatened filibusters more often, especially just recently. The combined 256 motions filed since the Democrats recaptured the Senate in the 2006 election is far more than the 130 motions Republicans filed in response to Democratic filibuster threats between 2003 and 2007 when George W. Bush was president and the GOP controlled the Senate (and since the current Congress is not over, presumably more are on the way).

These numbers are especially distressing when one considers that a 1975 reform, in recognition of the growing problem, was designed to reduce filibuster threats. Previously, sixty-seven votes (technically, two thirds of all voting senators, so possibly fewer) were required to invoke cloture and end debate on a question. The 1975 reform lowered the threshold to sixty (three fifths of all senators, so always sixty). And yet, the change produced more gridlock, not less. Koger writes:

These reforms institutionalized the notion that filibustering was an ordinary element of Senate decisionmaking. The 1975 reform may have lowered the threshold for cloture slightly (or not, depending on participation), but it also stabilized the threshold—thereby reducing the incentive to hold multiple cloture votes—while implicitly marking senators’ acceptance of supermajority rule in the Senate.

Somehow, it seems, a threshold of sixty-seven signaled to senators that the maneuver was to be used only in the most dramatic situations, but sixty sounds easier in theory for the majority to attain, and pressure from the minority on the majority to get sixty votes is somehow less unreasonable than pressure to get sixty-seven. Thus the practice of using cloture to end disputes became a routine part of Senate practice.

It was institutionalized for one more reason, in fact probably the main one. Minorities rarely pay a political price for threatening to filibuster. The cloture motion procedure is so arcane, so hidden from the view of all but the most initiated observers, that the public usually has no idea whom to blame for gridlock. When, for example, important executive branch positions go unfilled, average citizens tend to assume that the president simply hasn’t gotten around to appointing someone, whereas the truth is likely to be that the appointment was made months ago but has been languishing in the Senate (as of late May, 240 Obama appointees were awaiting confirmation votes). In 1993–1994, the Democratic majority filed eighty cloture motions, obliterating the previous record of fifty-nine and indicating numerous GOP threats of filibuster. The GOP not only paid no price in public disapproval; owing to the failure of Clinton’s health plan and other factors, it took back the Senate. The same, of course, could happen this November. In late June, a minority of forty-one, all but one of them Republicans, prevented a majority of fifty-seven from extending unemployment benefits to 1.2 million Americans. Many of those Americans know only that the Democrats are in charge in Washington and have no idea that a GOP minority blocked their payments. So Republicans may stand in unified opposition to Obama’s proposals, while also blaming Obama for being partisan; they may then benefit at the polls in November.

Koger’s study is highly academic and most readers will find Binder and Smith’s volume more accessible, even if it is now somewhat out of date. But Koger provides mountains of useful data, and his history comes to life in passages such as his account of Huey Long’s last filibuster. I’m not sure I agree, though, with his central conclusion, which is that the institutionalized acceptance of the sixty-vote Senate is a function of “the rising value of senators’ time.” That is, because senators are so pressed by so many time demands in our era, and because the Senate considers so many more matters than it did a century ago, the mere threat of a filibuster has led its members to conclude that “wasting time is more costly than accepting the outcome of a cloture vote,” even if the result isn’t to their liking.

I can readily believe that this is a factor, and perhaps an important one. It is surely one reason why we never see Mr. Smith–style filibusters anymore (another being, I think, fear by the minority of criticism in the press). But surely polarization is the main reason. If the current Republican caucus of forty-one senators included, say, twelve or fifteen moderates rather than the two to four who are there now, the demands on their time would be unchanged, but surely the health bill would have received a few GOP votes. The statistical formulas of the sort Koger marshals don’t point to polarization as the main answer; Senate watchers see evidence of polarization every day. Sarah Binder, who works at the Brookings Institution, believes that time constraints and polarization are linked: “Time is a constraint in part because minority members are making it hard for the majority to get things done,” she told me recently.

In any case, the Democrats are putting forth some interesting reform proposals. Tom Harkin of Iowa, who has been concerned about this issue for a decade or more, envisions a scheme to reduce the threshold whereby the first cloture vote would require sixty votes to end debate; a week later, if it failed, that would be lowered to fifty-seven; a week after that to fifty-four; and finally in week four to a simple majority of fifty-one. It’s a bit of a shell game, in that the majority can just wait the minority out, but it would give the minority a full month to make its case to the American public and change opinion.

Other possible approaches include redefining what is “debatable” under Senate rules. For example, under current Senate rules three motions are required in the Senate to go to a conference with the House intended to iron out the differences in the two chambers’ versions of a bill. Those motions are now subject to debate and sixty votes are needed to pass them. They could be declared “nondebatable.” Also, just as the Byrd rule freed budgetary items from cloture rules, the Senate could make other categories of measures—such as approval of judges and other presidential appointees—subject to simple majority votes. Finally, reforms could be declared to go into effect four or five Congresses from the time of voting, so neither side would know which party would immediately benefit.

Those interested in reform will face harrowing procedural hurdles. There are two ways to change the rules. One is a straightforward rewriting of Rule 22 itself. But—surprise—there’s a catch. When the senators of 1975 reduced the cloture threshold from sixty-seven to sixty, they retained the part of the rule stipulating that any future changes to the threshold would still require sixty-seven votes (a compromise that the reform faction, led by Walter Mondale, had to accept). That 1975 effort was bipartisan; Mondale worked with Republican James Pearson of Kansas. Such cooperation today, resulting in two thirds of senators agreeing to a rule change, seems utterly impossible.

The second route is championed by New Mexico Senator Tom Udall, who calls it “the Constitutional option.” Normally every odd-year January when Congress reconvenes, the House of Representatives declares itself a new body and must vote to readopt its rules; the Senate, however, declares itself a “continuing body,” simply rolling its rules over to the current term (not without justification, since two thirds of senators in any new Congress are not new). Udall wants the Senate to declare itself a new body next January. As such a body, it could change its rules by simple majority vote. Senate scholars appear to believe this is within the rules. But if we imagine the pandemonium that would be unleashed in the conservative press, especially after an election in which Democrats lost seats, it’s difficult to picture fifty-one Democrats holding together. And declaring the Senate a new body would make for open season on all rules, a fact of which the minority will surely be aware. Republicans, for example, may want to change rules to require supermajorities for tax increases, an idea that was embedded in Newt Gingrich’s Contract With America.

Reducing the supermajority threshold would, of course, benefit Republicans when they control the Senate. Partial privatization of Social Security, for example, might become more feasible (although it’s worth remembering that it was so unpopular in 2005 that no one even wrote a bill providing for it). But the current sixty-vote Senate makes the prospect of most serious liberal change unthinkable. What we have now, as Madison and Hamilton warned, is minority rule, with the Senate majority paying the political price for the gridlock as problems and crises go unaddressed. Unions, environmental groups, and liberals should confront the fact that it’s not a lack of presidential will that is the chief impediment to their agenda. They should mobilize accordingly.

—September 1, 2010

This Issue

September 30, 2010